Molecular determinants of pathogenicity and host specificity of highly pathogenic h5N1 BiH isolates

ABSTRACT Background: Towards preparation for a possible influenza pandemic, investigation of the molecular characteristics of the circulating avian H5N1 influenza virus strains is of crucial importance. These H5N1 viruses continue to spread, to infect animals and humans and to evolve and diversify providing so an ever-looming pandemic threat.Aim: To identify genetic structure and molecular biological characteristics of BiH's isolates of H5N1 HPAI as well as to assess the level of pathogenicity, phylogenetic origin and host- specificity of the isolates.Material and Methods: SPF embryonated chicken eggs were used for virus isolation. Viral RNA extracted using QIAamp viral RNA kit and manufacturer’s protocol (QIAGEN®) was used for PCR amplification. cDNA synthesis and PCR amplification of the coding region, using gene specific primer sets (primer sequences available on request), were carried out for all eight viral RNA segments separately. The Prism Big Dye Terminator v1.1 cycle sequencing kit (Applied Biosystems) was used and products were analyzed on an automatic ABI PRISM 3130 genetic analyzer (Applied Biosystems). Nucleotide sequences were analyzed using Bioedit software (v. 7.0.9.0) with an engine based on the ClustalW 1.4 algorithm. MEGA software (v. 4,0), using the neighbor joining tree inference analysis with the Tamura-Nei γ-model, was used to estimate phylogenies and calculate bootstrap values from the nucleotide sequences.Results: Full-length nucleotide sequences of the A/Cygnus olor/BIH/1/2006 (H5N1) strain were deposited in EMBL Nucleotide Sequence Database under accession nos. FN186008 to FN186014 and FM20943. The pathogenicity and host specificity of this strain, as polygenic traits, are determined in silico by the structure of its proteins, especially surface glycoproteins, HA and NA. Multibasic amino acid stretch PQGERRRKKR/GLF, marker of strains highly pathogenic to poultry, was present at the HA cleavage site of BiH strain. The RBS was typical for avian influenza viruses and contained Gln and Gly at positions 238 and 240 (H5 numbering) that is,226 and 228 according to H3 numbering with seven potential glycosylated sites but with increased binding to alpha2-6 sialoglycans thanks to substitutions, as follows, 110N, 171N, 171N, 172A, 205R and 251P. NA structure assigned this strain to the Z genotype, characterized also by the deletion of the five amino acid residues of the NS1 protein (positions 80-84). Amino acid residues, typical for the avian influenza viruses, were revealed in 40 out of 43 positions of M1, M2, NP, PA, PB2 and HA, determining the host range specificity. Phylogenetic analysis of the HA gene revealed that BiH isolates belonged to genetic clade 2.2., and presence of aspartic acid at the position of 403 of HA locate BiH isolates in 2.2.2. sub-clade.Conclusions: The BiH’s isolates were determined as HPAI virus with genes sequences closely related to A/Cygnus olor/Astrakhan/Ast05-2-10/2005 (H5N1). Three residues (M2 - 28V and 78K, NP - 33I), typical of human influenza viruses, were found, indicating a certain degree of intercurrent evolutionary adaptive changes in BiH isolates. Sequence comparison of HA and NA segments with relevant sequences in GenBank revealed that the BiH isolates and the ones from the southern Russia (Astrakhan region) group together phylogenetically, forming a monophyleticcluster in both genes indicating that these isolates have evolved from the same origin. Sequence derived phenotype markers of NA protein (E99, V129, D131, R136, H255 and Y256) as well as of M2 protein (26L, 27V, 30A, S31 and G34) showed that the isolates have an oseltamivir and amantadine sensitive genotype.